OpenAI internal design tasks - reported score
Professional Work
Release-reported internal design evaluation. Exact task cohort and grading are undisclosed; diagnostic.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-6 Astra | 50.0 |
Benchmarks
218 sources · 1,060 models × 2,130 benchmarks · 109,661 scoresProfessional Work
Release-reported internal design evaluation. Exact task cohort and grading are undisclosed; diagnostic.
| Model | Score |
|---|---|
| GPT-6 Astra | 50.0 |